Comparative study of K-Nearest Neighbor and Support Vector Machine methods in analyzing the consistency of college major based on high school majors

Fiorenza Rizkyllah, Anabel
Meiriza, Allsela
Hardiyanti, Dinna Yunika

Jurnal Sisfokom (Sistem Informasi dan Komputer)
Vol. 14 No. 4 (2025): NOVEMBER (In Press)

Choosing a college major is a crucial decision that can influence a student's academic and career path. Ensuring that students' choices are consistent with their high school majors can help improve academic success and career readiness. This paper delivers a comparative analysis of the K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) methods in evaluating the consistency of college major selection based on high school majors. A dataset of 636 students was collected and processed for analysis. The findings indicates that the KNN algorithm achieved an average precision, recall, F1-Score, and accuracy of 78%. Meanwhile, the SVM algorithm achieved a higher average score of 85%. This indicates better performance in analyzing the consistency between students' high school majors and their chosen college majors. These findings show that SVM is more effective in supporting guidance in college major selection, highlighting its suitability as a reliable method for decision making.